On Wed, 2005-08-10 at 22:05 -0500, William A. Mahaffey III wrote: > You could run the 32-bit Firefox & 32-bit plugins, they are *supposed* > to work seamlessly under the x86_64 OS. YMMV & all that. I have seen > much talk about this on the SuSE AMD64 list, and this recommendation has > floated out more than once. On Thu, 2005-08-11 at 05:30 -0500, Johnny Hughes wrote: > Right ... the only option would be to remove the x86_64 firefox and > install the i386 one ... but that might require MANY other i386 > libraries. (I can't test it here). > Tell them to get over it is another option :) This is _exactly_ what I do. I install Firefox i386 and many other components, including some of the Multimedia (like MPlayer) from the i386 pool. I ensure I _remove_ the x86_64 versions so YUM does not update them. This adds a few steps to auto-updating the i386 versions, but it's well worth it. I'm going to start playing with SmartPM when I have time to see if it can handle it's advanced form of Pinning so I can automate updates without having to do anything special. Of course, it would help if Red Hat shipped i386 packages for such Web components! > Should not be a huge performance issue ... at least I haven't noticed > any earth shattering performance enhancements between the x86_64 and > i386 distros when installed on x86_64 machines (that one could feel via > the GUI screen). AMD x86-64, as implemented in Athlon64/Opteron isn't so much about desktop performance differences. It's about the 1-4+GiB memory performance differences, I/O, etc... There are a few cases, especially if you optimize for the Athlon core's 3+3 ALU+FPU pipes, but other than that, it's really about the architecture and memory. -- Bryan J. Smith b.j.smith at ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman